Paperclip 📎 gives your agent access to millions of full-text biomedical papers. Agents like Claude Code, Cursor, and Codex are powerful research environments, not just for coding. But searching for papers often returns missing or incomplete results because these agents only have access to basic web search. With our release of Sy (sy.gxl.ai), we showed that giving agents access to a deep index of biomedical papers improves literature search (1.6× more accurate, 2.4× faster, and 3.6× cheaper) over Claude Code alone. 🚀 Today, we're releasing Paperclip as a CLI that can be installed directly in your agent IDE! 👉 To install and try it out, go to paperclip.gxl.ai 📖 We also wrote a technical deep-dive into how Paperclip works and why agents take to it really well: gxl.ai/blog/paperclip

Today, #paperclip is even faster and more reliable. We introduce Paperclip Repos, a git-like layer on our virtual filesystem that makes literature synthesis reproducible verifiable. This enables better grounding and large-scale scientific analysis gxl.ai/blog/paper-repos

Paperclip is the only "what does scientific literature say about XYZ?" skill I've tried that doesn't suck. Already had all of Pubmed {bio,med}rxiv 150 million OpenAlex abstracts... This is icing on the cake 😍

We added >220K FDA regulatory and >1M clinical trial docs to #paperclip. All natively indexed for agents and free. Now agents can easily reason over clinical studies w/o web search! E.g: find all trials that were approved despite missing endpoint gxl.ai/blog/adding-regulator…

huge gift to auto research community!

Very excited for this update!

Big Update🤩: #paperclip now includes full papers from all of arXiv, PubMed Central and 150 million abstracts!🖇️ You can give your LLM all that knowledge in one line—all optimally indexed for AI agents. Much more thorough and ~100x faster than web search, and free.

#paperclip now has a public repo! You can request features and discuss issues here github.com/GXL-ai/paperclip

been using the paperclip.gxl.ai/ and im just blown away but how much better it makes cc as a research assistant. i feel like they solved something smoothly. an elegant example of 'you need to translate problems to something that a CLI solves'. researchers - out there - try it @james_y_zou

As biology AI models mature, biology-native data infrastructure will be critical in separating the next generation of winning biotechs. It will come down to three things: biology-native data at scale, agentic AI across R&D workflows, and closed-loop lab automation.

You can now give your agent deep knowledge of millions of papers in one line with #paperclip!📎 >8 million papers natively indexed for agents. Much more thorough often 10x faster than standard deep research. Just add the paperclip mcp (instruction below).

A couple of exciting updates from the GXL team today: 1. Sy, our biomedical literature search agent, now has access to 8M papers from PubMed Central! Head over to sy.gxl.ai to try it out! 2. We’re releasing Paperclip, a CLI for biomedical papers! Agent IDEs like Claude Code, Cursor, and Codex are more than just for writing code; they are also powerful research environments. But the moment you need to find a paper, you're back in your browser, because the biomedical literature isn't easily accessible to the agent. Paperclip attaches a filesystem of millions of articles directly to your agent, so you can conduct literature search, pull full-text results, and synthesize findings without leaving your IDE. Search results are accessible locally, right alongside your work. We initially built Paperclip as an internal tool to power Sy, our biomedical literature search agent, but found that it works really well with general coding agents, so we’re making it available to other researchers to use. To install Paperclip: paperclip.gxl.ai Learn more: gxl.ai/blog/paperclip

Unlocking the unrealised potential of H&E-stained samples by AI generation of multiplex immunofluorescence 📷 Eric Wu et al Enable Medicine, CA, USA in @NatureComms ➡️ bpod.org.uk/archive/2025/10/… with @DrJohnAnkers

Does RAG solve hallucination? Even w/ RAG, we found that >30% of LLMs' medical statements are not fully supported by (sometimes contradict) the cited refs @NatureComms nature.com/articles/s41467-0… We present #SourceCheckUp agent to verify faithfulness of LM to source info. Great job @kevinywu @ericwu93 w/ awesome collaborators👏

Zou Lab launches MedArena, a free platform for clinicians to use and compare frontier LLMs MedArena is a free platform for clinicians to use and compare how frontier LLMs work on medical queries. Check it out at: medarena.ai/login

Hey #MedTwitter - Want to see how different LLMs handle your medical qs—for free? Check out MedArena 🏥, the first public arena where clinicians can evaluate and rank the latest medical LLMs. Go to medarena.ai to try it out! @ericwu93 @kevinywu @james_y_zou

Really cool way to evaluate LLMs for medicine in blinded fashion. It really helps build intuition on what works and doesnt. For example, I really like that there is RAG in some models (able to cite literature and references claims), but I've found when I've clicked into some of the references, it doesn't actually reflect the conclusion stated. Not sure if that's better or worse, since there seems be a bit of 'mode collapse' and for many prompts, the models all converge on similar (but incomplete) answers. Makes me wonder if they are mostly training on the same public text corpus (pubmed, google search, medscape, etc). @james_y_zou @kevinywu @ericwu93

Due to popular request, #medarena is now open to global clinicians! medarena.ai/login

Excited to share early updates from MedArena 🏥✨! 1⃣ Current frontrunner: @perplexity_ai 🤖, closely followed by Gemini 2.0 Flash Thinking 🌟—but the competition's just warming up! 🔥 2⃣ Great news: MedArena is global 🌍! Non-US clinicians are welcome—just provide your credentials at login. Feel free to share with clinicians to help us grow the leaderboard! medarena.ai 🚀 @james_y_zou @ericwu93 @EricTopol @NEJM_AI @David_Ouyang #MedTwitter #DigitalHealth #MedicalAI #MedTech